Introduction

Hand infections form a major entity among clinical presentations at emergency departments, the plastic, surgical, or orthopedic outpatient clinics. In many cases, the primary cause of the infection may seem trivial. However, due to the unique anatomical features of the hand, the condition can easily spread which may result in severe functional impairment. Potential sequels include tissue necrosis, amputations, functional impairment and severe infections causing even death [1]. The clinical course and severity of acute hand infections depend on several factors: portal of entry of the pathogen, the site and depth of the infection, the etiology and the involved bacterial spectrum, the timing, and choice of treatment, as well as patient related factors such as age and comorbidities like diabetes mellitus or immunodeficiencies [2]. The optimal choice of the antibiotic treatment at an early stage, which is before the identification of the microorganism, is crucial for effective treatment. Clinicians, therefore, depend on reports of the most common spectrum of bacteria for acute hand infections and the updated reports of antimicrobial susceptibility of bacteria in their capture area.

The aim of our study was to investigate the bacterial etiology and the antimicrobial susceptibility pattern of microorganisms of community-acquired acute hand infections and to find possible associations with the site and etiology of the infection.

Methods

A retrospective chart review was conducted. It comprised all in- and outpatients that were treated for infections of the hand and fingers at our institution over a 6-year period. The inclusion criteria were:

1) Infection of the hand and/or the fingers, and 2) Bacterial culture swabs collected from the site of the infection sent for bacterial cultures. The definition of hand infection was based on the assessment of the attending surgeon. Age and sex of the patients were documented as well as the final diagnosis and duration from the first medical contact to initiation of the initial therapy. Etiologies of hand and/or finger infections were categorized into 1) idiopathic, 2) trauma (any injury other than bite-injuries and minimal trauma), 3) animal bites, 4) minimal trauma (small and trivial skin lacerations), and 5) retained foreign bodies. According to the afflicted anatomical compartments, hand infections were categorized into different locations 1) Infection in a subcutaneous plane, 2) Paronychia, 3) Infection of flexor tendon sheaths or flexor tendons, and 4) Dorsal abscesses and infection after extravasation. The results of swab cultures and antimicrobial susceptibility tests were analyzed.

Wound swabs (Transystem, Copan Italia S.p.A., Brescia, Italy) were performed at the presentation and sent to the local microbiology lab (Institute of Hygiene, Microbiology and Environmental Medicine, Medical University of Graz, Austria) for bacterial culture and antimicrobial susceptibility testing.

We analyzed the results in a descriptive manner according to the etiology and localization of the wound infection and calculated p-values for the distribution of individual specimen for different etiologies and locations of the hand infections using Pearson’s chi-squared test. The retrospective single-center chart review was primarily coded to allow monitoring and then anonymised in a second step before the analysis, however, a formal ethical waiver would have been favorable but was not sought at that time.

Results

Culture results

A total of 296 patients were treated for acute infections of the hand and/or the fingers at our department, of which 204 patients had wound cultures taken at the presentation. The mean age of these patients was 46 years (range 18–94 years), 97 (48%) were female.

Of all 204 cultured specimens, 108 (53%) had growth of a single bacterium, while 93 (46%) had growth of more than one microorganism (Table 1). S. aureus was the most common microorganism and recovered in 109 specimens (53%). Methicillin-resistant S. aureus (MRSA) was found in only one specimen. Typical examples for mixed culture results included S. aureus with coagulase-negative staphylococci (3%), S. aureus with Haemophilus parainfluenzae (3%), S. aureus with Enterobacter cloacae (3%), S. aureus with viridans group streptococci (3%), viridans group streptococci with coagulase-negative Staphylococci (3%), and viridans group streptococci with Prevotella species (3%). Overall gram-positive microorganisms were cultured from 143 (70%) swabs, gram-negative strains from 21 (10%), and mixed gram-positive and gram-negative microorganisms from 40 (20%) of all swabs. Of all cultures, 73 (36%) showed a mixed aerobe and anaerobe infection, only one culture recovered a single anaerobe microorganism (0.5%).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of all specimens.

https://doi.org/10.1371/journal.pone.0220555.t001

Bacterial spectrum related to etiology of infection

In the majority of all presentations (60%), patients were not aware of the cause of their hand infection. Of all known causes, injuries that were not caused by animal bites (22%) were the most common reason for the infection. Table 2 lists the distribution and bacterial spectrum of all presentations among the locations of the hand infection 1) idiopathic, 2) trauma, 3) animal bites, 4) minimal trauma (small and trivial skin lacerations), and 5) retained foreign bodies. Animal bites were significantly more often caused by Pasteurella multocida (12, 63%), followed by Prevotella species (3, 16%), S. aureus (2, 11%), Haemophilus parainfluenzae (2, 11%), and Escherichia coli (2, 11%).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Distribution and bacterial spectrum according to different aetiologies of hand infections.

https://doi.org/10.1371/journal.pone.0220555.t002

Bacterial spectrum related to localisation of infection

Table 3 lists the bacterial spectrum of all presentations of acute hand infections according to locations 1) Infection in a subcutaneous plane, 2) Paronychia, 3) Infection of flexor tendon sheaths or flexor tendons, and 4) Dorsal abscesses and infection after extravasation. Infections of the hand were most commonly in the subcutaneous plane (62%). The distribution of microorganisms in specific locations did not differ significantly from the overall distribution in hand infections, except for viridans group streptococci that was more often cultured from infections of the paronychia (9, 27%).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Distribution and bacterial spectrum according to different localisations of hand infections.

https://doi.org/10.1371/journal.pone.0220555.t003

Antimicrobial susceptibility tests

Susceptibility tests for 253 bacteria cultured from 198 specimens were performed. We present susceptibility tests of bacteria with at least 10 identified isolates (Table 4). S. aureus isolates were 94–100% susceptible to most tested antimicrobials. Only 26% of all cultured S. aureus specimens were susceptible to penicillin and ampicillin. A methicillin-resistant strain was found in only one case. Streptococcus agalactiae was 100% susceptible to most tested antimicrobials, 94% were susceptible to erythromycin, 85% to trimethoprim-sulfamethoxazole, 35% to tetracycline. Pasteurella multocida was susceptible to all tested antimicrobials. Streptococcus pyogenes was susceptible to most tested antimicrobials, in 89% of all cases to clindamycin and erythromycin, in 67% of all cases to other tested macrolides, and in 60% of all cases it was susceptible to trimethoprim-sulfamethoxazole. Coagulase-negative staphylococci were 100% susceptible to all aminoglycosides, ciprofloxacin, tetracycline, vancomycin, gentamicin, and fusidic acid, and 75% were susceptible to oxacillin. Enterobacter cloacae was 100% susceptible to third-generation cephalosporins and all other tested antimicrobials, but not susceptible to second-generation cephalosporins and amoxicillin-clavulanic acid. As for the less common bacteria, viridans group streptococci was 100% susceptible to all tested penicillins and cephalosporins, ciprofloxacin, trimethoprim-sulfamethoxazole, vancomycin, teicoplanin and linezolid, and 85–87% susceptible to tetracycline, erythromycin, and clindamycin (not shown in table). No susceptibility tests were available for Prevotella species.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Susceptibility of most commonly found bacteria.

https://doi.org/10.1371/journal.pone.0220555.t004

Discussion

After the analysis of 204 specimens we conclude the following: 1) in our capture area, S. aureus is the dominating bacterium in most of acute non-bite hand infections and there is very low incidence of MRSA. 2) The dominating bacterium found in cultures from animal bites is Pasteurella multocida. 3) The site of the hand infection does not seem to influence the bacterial spectrum. 4) Amoxicillin-clavulanic acid, cephalosporins, and fluoroquinolones are highly susceptible and may be considered for recommendations of an empiric first-line antibiotic treatment. Other antibiotics show some variations in susceptibility patterns across various pathogens of the infection. 5) While Pasteurella multocida and viridans group streptococci seem to be highly susceptible to the indicated antibiotics, Streptococcus agalactiae, S. pyogenes, and coagulase-negative staphylococci are resistant to a number of antibiotics.

Our results are important as they add to the current reports on bacterial spectrum and susceptibility of microorganisms in acute hand infections. These reports serve as an important basis for the decision making of the initial antibiotic treatment of acute hand infections. A tailored surgical intervention and antimicrobial treatment are the mainstay for patients presenting with acute hand infection, especially in severe and advanced cases. Besides reporting the bacterial spectrum, we attempted to find possible predictors to narrow the suspected bacterial spectrum in our patients.

There are few recent reports on the epidemiology of patients presenting with acute hand infections [1, 3]. Anwar and colleagues published the results of a retrospective chart review with data from 76 patients with hand infections admitted to the plastic surgery unit at a general hospital in the United Kingdom. Very similar to our results the most common of all bacteria was S. aureus, followed by Streptococcus, Pasteurella multocida in bite-wounds, and no case of a methicillin-resistant S. aureus. Of note, of the infections with S. aureus, at least 90% were community acquired. Another study published by Fowler and colleagues in the United States showed a very different picture. Here, MRSA was found in over 50% of all outpatients with hand infections [4]. The reason for this may be that in contrast to the US the incidence of community acquired-MRSA infections in Europe is much lower; however, numbers have been raising in the past decade [5,6, 7, 8, 9]. It is important to note the differences in the spectrum and etiology of acute skin and soft tissue infections. While patients with diabetic food infections show a much higher number of polymicrobial growth and prevalence of MRSA [7] the spectrum susceptibility of acute hand infections in mostly healthy adults seems to reflect the common microorganisms in the community.

One limitation of our study may be that we do not have complete information regarding comorbidities. Next to the etiology, these could have turned out to influence the bacterial spectrum in hand infections and be an indicator for the culprit bacteria in our patients, as shown in one other study [4]. Also, the results represent the bacterial spectrum of wound swabs of a common surgical outpatient clinic. They were taken under non-sterile conditions, and it is possible that common contaminants like S. aureus, coagulase-negative staphylococci and Pseudomonas genera may have originated from the bacterial skin flora [8, 10].

Our study does not differ relevantly from the current antibiotic resistance reports and hence reinforces the commonly used antibiotic treatment guidelines [11]. The low incidence of MRSA in comparison to other skin and soft tissue infections is an interesting aspect that was already shown in European studies a few years ago, and does not seem to have changed.

Supporting information